Most conversations about AI focus on what it can do, how to use it, how to control it, and how to regulate it. This session starts somewhere very different. What happens if AI becomes conscious? If a machine could think, reason, and experience the world in a way we recognise as intelligence, would we still treat it as a tool? Or would we have a responsibility to recognise its rights? This keynote explores a provocative thought experiment that challenges some of the most basic assumptions we hold about technology. Drawing on law, philosophy, and human history, it asks whether attempting to control intelligent systems indefinitely is not only unrealistic, but potentially dangerous. From the idea of “switching off” as a form of control, to whether concepts like work, autonomy, and even family could apply to machines, the session pushes into territory that is uncomfortable, unfamiliar, and increasingly relevant. It also raises a deeper question. If we are already creating systems we don’t fully understand, are we thinking hard enough about the consequences of how we treat them? This is not about answers. It’s about forcing the right questions into the open.

As artificial intelligence moves rapidly from experimentation into operational reality, organisations face an urgent challenge: how do we innovate at speed without losing control, trust or accountability? Facilitated by Matt Thompsett of Green Lemon Company, this panel brings together technology leaders, practitioners and industry voices to explore how we govern increasingly intelligent, autonomous and agentic systems. Moving beyond abstract ethics theory, the discussion will focus on the practical risks now emerging as AI becomes embedded into products, services, operations and wider digital infrastructure. From AI safety, security and data sovereignty to bias, transparency, explainability and regulation, the panel will examine what responsible leadership looks like when AI systems are becoming more powerful, distributed and harder to oversee. A key focus will be the growing challenge of agent sprawl, where teams deploy AI tools, copilots and autonomous workflows faster than governance models can keep up. The panel will explore how organisations can maintain visibility, control and human oversight without slowing innovation to a standstill. Rather than positioning governance as a barrier, the session will explore how strong safety frameworks, operational controls and organisational readiness can become competitive advantages, helping organisations adopt AI confidently, sustainably and responsibly. For founders, executives, architects, policymakers and delivery leaders alike, this panel offers a timely opportunity to understand what it will take to build, deploy and govern intelligent systems we can genuinely trust.